Accredited Standards Committee X3, Information Processing Systems Doc. No.: X3T9.2/92- R0 Date: November 12, 1992 Project: Ref. Doc.: Reply to: J. Lohmeyer To: Membership of X3T9.2 From: Lawrence J. Lamers, Secretary X3T9.2 John Lohmeyer Chair X3T9.2 Subject: Minutes of General November 11-12, '92 Working Group Meeting ______________________________________________________________________________ Agenda 1. Opening Remarks 2. Attendance and Membership 3. Approval of Agenda 4. SCSI-3 Command Sets 4.1 Synchronization Status and Errors (92-130R2) [Robison] 4.2 Cache Control Revisited (90-21R4) [Milligan] 4.3 Master Target Save Disable Bit (92-218r0) [Milligan] 5. SCSI-3 Architecture Model 6. Serial Protocol Topics 6.1 SCSI-3 Serial Bus Protocol (92-199R2) [Marazas] 6.2 SCSI-3 Fiber Channel Protocol (FCP) 6.3 SSI-3 Generic Packetized Protocol (GPP) 7. Directly-Addressable Device Interface 7.1 Presentation on P1285 work [Chris Hamlin] 8. General Working Group Topics 8.1 Proposal to add disconnect unconditionally (92-133R2) [Smyers] 8.2 Conformance Testing [Stephens] 8.3 Report on CAM-2 Study Group Meeeting [Dallas] 8.4 Where does SPI really fit in SCSI? (92-216r0) [Ham] 9. SCSI-3 Parallel Interface (92-010r8) [Spence] 9.1 Proposal for Input Leakage Current Specification (92-219r0) [Asami] 9.2 50-line to 68-line connections (92-215r0) [Ham] 9.3 Suggested Changes to Figure 8 (92-220r0) [Goldie] 9.4 Pin Capacitance Measurement [Wallace] 9.5 Proposed Cable Preparation Procedure for Twisted Pair (92- 210) [Debiec] 9.6 Proposal for SPI Cable test Procedures (92-150) [Debiec] 10. Meeting Schedule 11. Adjournment Results of Meeting 1. Opening Remarks John Lohmeyer the Chair, called the meeting to order at 9:00 a.m., Tuesday, November 11, 1992. He thanked Rich Taborek of Amdahl for hosting the meeting. As is customary, the people attending introduced themselves. A copy of the attendance list was circulated for attendance and corrections. It was stated that the meeting had been authorized by X3T9.2 and would be conducted under the X3 rules. Ad hoc meetings take no final actions, but prepare recommendations for approval by the X3T9.2 task group. The voting rules for the meeting are those of the parent committee, X3T9.2. These rules are: one vote per company; and any participating company member may vote. The minutes of this meeting will be posted to the SCSI BBS and the SCSI Reflector and will be included in the next committee mailing. 2. Attendance and Membership Attendance at working group meetings does not count toward minimum attendance requirements for X3T9.2 membership. Working group meetings are open to any person or company to attend and to express their opinion on the subjects being discussed. The following people attended the meeting: Name Status Organization Phone Number ------------------------- ------ ------------------------------ -------------- Mr. Norm Harris P Adaptec, Inc. (408) 945-8600 x2230 Mr. Jan V. Dedek P Ancot Corp. (415) 322-5322 Mr. Gary Porter V Ancot Corp. (415) 322-5322 Mr. Scott Smyers P Apple Computer (408) 974-7057 Mr. Clifford E. Strang P BusLogic (408) 492-9090 Jr. Mr. Bob Gannon O C&M Corp. (203) 774-4812 Mr. Geoffrey Hyman O Cables To Go (513) 275-0886 x319 Mr. Nicos Syrimis A Cirrus Logic Inc. (510) 226-2153 Mr. John Geldman A Cirrus Logic Inc. (510) 226-2368 Mr. Boyd G. Pett V Cirrus Logic Inc. (510) 623-8300 Mr. Anwar Khan O Cypress Semiconductor (408) 943-2574 Mr. Michael Smith S Dallas Semiconductor (214) 450-0457 Mr. Tom Hunt V Dallas Semiconductor (415) 617-1790 Mr. Charles Monia P Digital Equipment Corp. (508) 841-6757 Mr. William Dallas A Digital Equipment Corp. (603) 881-2508 Dr. William Ham A Digital Equipment Corp. (508) 841-2629 Mr. Edward A. Gardner A Digital Equipment Corp. (719) 548-2247 Mr. Ralph Weber A Digital Equipment Corp. (603) 881-1497 Mr. Skip Jones P Emulex Corp. (714) 668-5058 Mr. Tingli Chan V Emulex Corp. (714) 668-5487 Mr. I. Dal Allan P ENDL (408) 867-6630 Mr. D. W. Spence A ENDL Associates (512) 255-0339 Mr. Colm Tracey O Fujikura America, Inc. (404) 956-7200 Mr. Joel Urban O Fujitsu Microelectronics (408) 922-8928 Mr. Jeffrey L. Williams P Hewlett Packard Co. (208) 323-5030 Mr. Kurt Chan A Hewlett Packard Co. (916) 785-5621 Mr. Yoshihiko Yano O Hitachi Computer Products (408) 986-9770 Mr. Howard Wang O Hitachi Computer Products (408) 986-9770 x207 Mr. George Penokie P IBM Corp. (507) 253-5208 Mr. Gerald Marazas A IBM Corp. (407) 982-4423 Mr. Gary R. Stephens A IBM Corp. (602) 799-2246 Mr. Giles Frazier S IBM Corp. (512) 838-1802 Mr. Tom Taetsch O Intellistor, Inc. (303) 682-6651 Mr. David Lawson P Interphase Corp. (214) 919-9000 Mr. Rick Paul O Judd Wire, Inc. (619) 744-7720 Mr. Raymond C. Yule P Micropolis Corp. (818) 718-7803 Mr. Robbie Shergill P National Semiconductor (408) 721-7959 Mr. John Goldie A National Semiconductor (408) 721-2075 Mr. Manish Gupta O National Semiconductor (408) 721-5728 Mr. John Lohmeyer P NCR Corp. (719) 596-5795 x362 Mr. Rod DeKoning S NCR Corp. (316) 636-8842 Mr. Erich Herter V NEC Germany 49-89930065 Mr. Dennis P. Trupski O Olivetti (408) 366-3254 Mr. Stephen F. Heil P Panasonic Technologies, Inc (201) 348-7064 Mr. Gene Milligan A Seagate Technology (405) 324-3070 Mr. Forrest Crowell P SGS-Thomson Microelectronics (714) 957-6018 Mr. Stephen G. Finch P Silicon Systems, Inc. (714) 573-6855 Mr. Robert N. Snively P Sun Microsystems, Inc. (415) 336-5332 Mr. Vit Novak A Sun Microsystems, Inc. (415) 336-2455 Mr. Todd Sprenkle V Tandem Computers (408) 285-6264 Mr. Kevin Gingerich V Texas Instruments (214) 997-3370 Mr. Alan Ebright P Toshiba America, Inc. (408) 428-4831 Mr. Jeff Stai P Western Digital (714) 932-7644 53 people present Status Key: P - Principal Member A - Alternate Member L - Liaison Member O - Observer S,V - Visitor 3. Approval of Agenda The proposed agenda was approved. 4. SCSI-3 Command Sets 4.1 Synchronization Status and Errors (92-130R2) [Robison] Rod DeKonig presented a proposal regarding the reporting of spindle synchronization. The proposal includes a new diagnostic page to report drive status regarding the state of spindle synchronization. The new pages have a large amount of reserved space (40 bytes) plus seven bits of information. The working recommends acceptance of this proposal for SCSI-3. 4.2 Cache Control Revisited (90-21R4) [Milligan] Should the segment size be in bytes or logical blocks? Currently it is in bytes so the limit is 64kbytes reported as the segment size. John Lohmeyer suggested adding a bytes or blocks bit. This should preserve the existing SCSI-3 implementations. Gene Milligan agreed to document this addition. Later in the meeting, Gene brought in document 92-217. It defines a LBCSS (Logical Block Cache Segment Size) bit which controls whether the segment is reported in bytes or blocks. The working group recommends that X3T9.2/92-217 be accepted for inclusion in SCSI-3. 4.3 Master Target Save Disable Bit (92-218r0) [Milligan] This proposal adds a bit to the CONTROL MODE page to overide the TSD bit in the individual log pages. The term "Master" will be changed to "Global Log". This saves an initiator from sending each individual log page to control the saving operation. The proposal as revised is recommended for inclusion in SCSI-3. 5. SCSI-3 Architecture Model Charles Monia presented the latest concepts on SAM. The terminology now speaks of executives, managers, and agents. The new terminology did not sit well with everyone and Charles will head off and do some more work. 6. Serial Protocol Topics 6.1 SCSI-3 Serial Bus Protocol (92-199R2) [Marazas] Gerry Marazas presented the latest serial bus protocol for IEEE P1394. He stepped through an illustration on the use of tap packets. He illustrated the command packet. There are two outbound FIFOs, command and urgent, and one inbound FIFO for status. There was some concern about having 12 bytes for the status block and also having information that is redundant with the sense buffer. Discussion on dealing with resource limitations (i.e., busy conditions). Handling aborts - as proposed abort message does not get a response (there is a transport level acknowledge). Therefore it is not equivalent to the current thinking in SAM, because the resource is not immediately available following the acknowledge. Some form of confirmation service is needed. Jerry Marazas agreed to address the issue, which also exists for clear queue. There is a one-to-one corelation between command blocks and status blocks. There are no status blocks for unacknowledged services. There was a debate over whether we should limit SCSI-3 CDBs to twelve bytes vs. sixteen bytes. In particular, are the reserved bytes in the SBP command block reserved for SBP or reserved for SAM? The issue was elavated for a plenary vote. Jerry listed several issues and their status: round robin fetch - open order vs simple - open initiator_id - open dynamic command append - open command block revisions - open abort flag protocol - settled justification for chaining - open An editors session is planned for November 23-24, 1992 in Colo Spgs, CO at Digital -- Contact Ed Gardner (719-548-2247) for details. 6.2 SCSI-3 Fiber Channel Protocol (FCP) Bob Snively elaborated on the recent changes in FCP. a) initiator/target sequence designation corrected b) link application frames for ULP c) Exchange abort -> SCSI aborts; remove resets d) FCP Xfer Rdy optionally disabled (uses link application frame) e) remove luntar Bob plans to remove the BUS DEVICE RESET message function and he may add a priority override reservation. If there is no reset function, then other changes may be necessary including an expiration on I/O processes. There are probably other impacts that will affect the device drivers. There has been a lot of pressure to remove the need for Xfer Rdys for the FC- LT. However, the initiator must be able to deal with a target buffer overrun. There is problem with the SEND command because the AEN address field size is not big enough to handle the expanded LUN fields. [ljl: Perhaps the SEND command should be moved to the SPC document.] Bob is looking into accomodating the IEEE P1275 open boot standard in FCP. 6.3 SSI-3 Generic Packetized Protocol (GPP) Deferred. 7. Directly-Addressable Device Interface 7.1 Presentation on P1285 work [Chris Hamlin] Chris Hamlin was not available. David James made a presentation on RAM-LINK. RAM-LINK has architectural characteristics of one layer that is similar to the P1285 beta layer. RAM-LINK is being developed as part of the IEEE P1596 project. 8. General Working Group Topics 8.1 Proposal to add disconnect unconditionally (92-133R2) [Smyers] Deferred. Larry Lamers and Scott Smyers plan to work on this document and bring it into the San Diego. 8.2 Conformance Testing [Stephens] Deferred. 8.3 Report on CAM-2 Study Group Meeeting [Dallas] Bill Dallas reported. CAM needs a broader scope; however an investigation of the compatibiity with CAM-1 needs to be done. Binary compatibility is not likely to be achieved. Sixty-four bit systems break compatibility, while thirty-two bit systems may be compatibile. They plan to investigate callbacks as the method of XPT and driver notification vs. allowing additional methods (e.g., mailboxes, messaging). The driver portability amoung operating systems is 80% and amoung platforms is 95-100%. They plan to investigate using annexes to permit operating systems to define structure offsets and to permit CAM-2 to use named fields instead of bit fields. A survey of third-party vendors is needed before taking this step. It is intended that CAM-2 will support SCSI-3 functionality. 8.4 Where does SPI really fit in SCSI? (92-216r0) [Ham] How does one identify a mixed-implementation device? A taxonomy of compliance is needed to help marketing folks identify their devices. The introduction part should give an explanation of the interelationship of SCSI- 2 and SCSI-3. Bill will attempt wording on this. John Lohmeyer also believes a press release is needed. 9. SCSI-3 Parallel Interface (92-010r8) [Spence] 9.1 Proposal for Input Leakage Current Specification (92-219r0) [Asami] Jeff Stai presented Tak Asami's proposal on input leakage current. The leakage current specified in SPI is not intended for hot-plugging events. At the initial insertion there is a peak current draw until a steady-state condition is established. There were a number of questions that only Tak could answer so the consensus was to postpone further discussion until the San Diego meeting in December. 9.2 50-line to 68-line connections (92-215r0) [Ham] Bill Ham is striving to have a single transition device for 68- to 50- conductor cables. This entails connector conductor 25 in the A cable to TERMPWR instead of leaving it open. There was also considerable discussion on interconnecting 8-bit and 16-bit devices. If the bus is continous and a high-density connector is used for connecting the 8-bit device to the P cable there is agreement that it works. When making a transition of 16-bit bus to an 8-bit bus there is not a lot of agreement. In the final analysis we didn't change anything except to add a note to the figures to clarify that the numbers are conductor numbers, not connector contact numbers. 9.3 Suggested Changes to Figure 8 (92-220r0) [Goldie] RS-485.1983 is modeled for 4000 ft. without a ground wire allowing for 32 devices and 7 volt ground shift. This model does not refect the way they are used in an SCSI application. John Goldie suggested changing the test circuit to more closely follow the actual SCSI application requirements. John proposed a new test circuit that models the SCSI bus with 32 devices and eliminates the DC common mode voltage. This test guarantees the signal swing of 1 volt. An undriven line will be at 800 mv with 32 devices. In absense of any objection, Figure 8 will be revised. John also proposed to eliminate item a) in 7.2.3 because this requirement is already stated in RS-485 and is redundant. This proposal was accepted Gene Milligan stated that the requirements letter transmitted to ISO should be revised to reflect these changes. John Goldie agreed to attempt to get a copy of the revised 485 proposal for X3T9.2 review. 9.4 Pin Capacitance Measurement [Wallace] Ting Lee stated that commercial capacitance meters are at 5 Mhz. The bias voltage during the measurement is a critical factor that can drastically change the capacitance. The filtering used for glitch protection also affects the capacitance measured. The consensus was to measure capacitance at: zero bias, no voltage, 100 mv peak-to-peak amplitude, 5 Mhz, contacts not connected. 9.5 Proposed Cable Preparation Procedure for Twisted Pair (92-210) [Debiec] This item will be covered at the next meeting. 9.6 Proposal for SPI Cable test Procedures (92-150) [Debiec] This item will be covered at the next meeting. 10. Meeting Schedule The next meeting of the General Working Group will be January 12-13, 1993 at the Sheraton Hotel (408-745-6000) in Sunnyvale, CA hosted by Tandem Computers. The room rates are $100.00 plus tax. The reservation deadline for these rates is December 18, 1992. 11. Adjournment The meeting was adjourned at 5:30 p.m. on Wednesday November 11, 1992.